[SE/Linux] status / progress report 13jun2004

From: Luke Kenneth Casson Leighton (lkcl_at_lkcl.net)
Date: 06/13/04

  • Next message: Chris Metzler: "Re: Gnome 2.6 and MAC OSX-like panel"
    Date: Sun, 13 Jun 2004 15:36:48 +0000
    To: debian-devel@lists.debian.org, SE-Linux <selinux@tycho.nsa.gov>, debian-user@lists.debian.org
    
    

    This is a status / progress report for Debian / SE/Linux integration.
    I look forward to the day when it need no longer be maintained,
    which will be when all of the outstanding issues have been addressed.

    The constant work-in-progress version of this report will always be
    available from:

            http://hands.com/~lkcl/selinux

    The major outstanding issues are:

    * debian kernels need to be available compiled with se/linux security
      enabled (and boot-time optional) by default. this results in a
      2% performance hit (wow big deal) when se/linux is not enabled
      at boot time. Gentoo, SuSE and Fedora all accept this 2%.

    * sarge freeze is holding back libselinux1 from being made "Required"
      which is holding pretty much evveerrything up, but there is a
      temporary idea (do a package se-<pkgname>) as a workaround.

    * a decision needs to be made on dpkg either to accept the postinst.d
      idea or come up with a workable alternative. decision appears to
      be held up because people "don't like the idea of selinux" rather
      than for any genuine technical reason.

      "alternative" patched dpkg package that provide the postinst.d
      functionality will be made available "ad infinitum" until a
      decision is made.

      ... how about an se-dpkg? maybe the se_apt-get, se_dpkg,
      se_dpkg-reconfigure scripts could be moved into it, at the
      same time?

    * the idea of using a pam_selinux.so for everything has been disrupted
      slightly for certain packages such as kdm, openssh, because the
      ordering of opening ttys and calling the pam session stuff tends
      to be moved about by upstream developers - without consideration
      as to the impact it will have. pre-pam_selinux patches (esp. for
      openssh) have been "dusted off".

    * pam seems to have "lost the plot" a bit and serious consideration
      is being given to doing a fork for BOTH redhat AND debian.

      [the debian pam maintainer has a staggering FIFTY upstream
       patches in debian/patches/ for 0.77. he's prepared to accept
       ANOTHER patch to add to the list, for selinux, but only
       against latest cvs - 0.78 or above. redhat also have to
       maintain their own patches - against 0.76 - which includes
       bug fixes that aren't in the "alternative" debian packages
       yet, and it's all just going pear-shaped]

    packaging:

            * "alternative" unstable packages (which had had to be patched,
              see individual status reports below) for:

                    coreutils, cron, dpkg, init, kern, logrotate and pam
            
             are all available from http://selinux.lemuria.org/newselinux
             (or from the original http://www.coker.com.au/newselinux)

            * "standard", or "default" packages for unstable (sid)
            
                    selinux-policy-default, selinux-utils, libselinux1,
                    checkpolicy, policycoreutils and selinux-doc
            
              are available from the debian mirrors - current versioning
              is 1.12-2 to 1.12-3 of these packages.

    NSA/SELinux kernel 2.6:

            http://www.nsa.gov/selinux/code/download5.cfm
            http://sf.net/projects/selinux/ (see cvs).

            status: most of the selinux enhancements are available
                    upstream in 2.6, however the very latest patches
                            are only available from the above sites.

    debian:

            http://lists.debian.org/debian-devel/2004/05/msg01738.html

            status: presently, base packages are frozen and no modifications
                    or additional packages are allowed (to base). this
                            affects libselinux1 status from being changed, and therefore
                            pretty much everything else from thereon down.

                            temporary measure idea for maintainers is to produce
                            "se-pkgname" which will later on be an empty package
                            depending on "pkgname".

    debian kernel 2.6 images:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=249510
            http://open.hands.com/~lkcl/selinux

            status: raised only 12 days ago. requested that se/linux
                        security config options be enabled in stock
                        Debian kernels but require selinux=1 and enforcing=1
                            to switch it on.

    coreutils:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=193328

            status: 1 year old, requested information, information now
                    provided, upstream and maintainer prodded for
                            acknowledgement. [30may2004] mike stone responded
                            by saying that it's unlikely that action will be taken
                            until after sarge is released.

    logrotate:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=224880

            status: russell alerted maintainer that upstream inclusion
                    is done (157 days ago) but debian package 3.7-1
                            disables it by default due to libselinux1 not being
                            "base/required" or "important". change made to
                            libselinux1 to reflect that.
                            
                            [30may2004] paul martin confirmed that he is waiting
                            for this change, and the "ftpmasters" need to make
                            the decision.

                            13jun2004: pinged paul suggesting the se-<pkgname>
                            idea.

    cron:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=193644

            i think this one's my favourite.

            status: 1 year old. bit of a wing-ding and misunderstanding
                    over a field name, fortunately the maintainer stood
                            his ground until the non-cron-code-experts understood
                            the issues. updated patch sent.
                            31may2004: steve (maintainer) evaluating patch. also
                            steve aware of sarge freeze and implications.
                            8jun2004: bug found in cron which was accidentally
                            fixed in selinux version. steve (maintainer) now
                            happy. to check / confirm latest patch with sds (nsa)
                            8jun2004: steve to create a cron and se-cron package
                            where se-cron will be a dummy package when sarge
                            is released (and libselinux1 goes to "Required").

                            10jun2004: dan walters created new patch, with some
                            additional cleanups etc. sent to steve (maintainer)

    pam:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=249499
            http://www.redhat.com/archives/pam-list/2004-May/msg00058.html

            status: amazingly, only 19 days old. unless there's an
                    earlier one and it's already been integrated
                            upstream. changes are only to pam_unix, apparently,
                            on that one (and there's another patch for pam_selinux).
                    information sought from upstream and from the
                            maintainer.
                            30may2004: several messages to upstream explaining
                            that pam_selinux.so is needed upstream before
                            other packages can start putting
                            "session required pam_selinux.so" into upstream
                            as well.
                            30may2004: subscribed direct to list to avoid
                            moderation and wrote message explaining situation
                            (pam upstream acceptance or lack of equals major
                             hold-up).
                            1jun2004: issue with packages opening and closing
                            sessions, plus upstream packages moving the place
                            where pam is called from (e.g. openssh) causing
                            tty problems. serious consideration being given
                            to reinvoking / dusting-off the selinux patches that
                            pam_selinux was supposed to do away with, on the
                            basis that upstream authors are less likely to
                            interfere with the ordering of "#ifdef WITH_SELINUX"
                            than they are with moving calls to pam_open_session.

                            8jun2004: situation with pam is bad: no communication
                            whatsoever received from upstream. bugs in 0.76 fixed
                            for fedora, too much work to back-port. serious
                            consideration being given to forking pam. debian
                            maintainer happy to accept patch against latest sf.net
                            cvs (0.78 or above)

    dpkg:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=249496
            http://lists.debian.org/debian-dpkg/2002/03/msg00154.html
            http://lists.debian.org/debian-devel/2002/03/msg02063.html
            http://lists.debian.org/debian-dpkg/2004/05/msg00255.html
            http://lists.debian.org/debian-devel/2004/06/msg00698.html

            status: mr russell coker's postinst.d patch is apparently
                    well-known and the bugreport has been merged with
                            other bugs, one of which (#17243) dates back to
                            1998! kuudosss. however, the maintainer says that
                            those bugs are part of a larger picture of
                            required / requested functionality and they don't
                            want to proceed with what would turn out to be a
                            temporary measure.

                            30may2004: after evaluating options (see links
                            above) initiated thread to convince dpkg
                            developers to incorporate postinst.d patch.

                            13jun2004: no response yet received, another ping
                            initiated.

    init:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=242900

            status: raised 50 days ago. seeking information from
                    debian maintainer.

                            13jun2004 contact. advised maintainer of
                            se-cron idea pending sarge unfreeze, suggested
                            doing an se-init (se-sysvinit), temporarily.

    openssh:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=193664

            status: 30may2004 - russell's explained that this patch is no
                    longer needed because the patches to PAM deal with
                            this, now.

                            8jun2004 - serious consideration being given to
                            requesting the (retired) openssh WITH_SELINUX
                            patch be added due to calls to pam_open_session
                            having been moved to before ttys are set up
                            (in sshd). it's all gone pear-shaped.

                            10jun2004: investigation by dan and russell leads
                            to a decision to reintroduce the former openssh patch,
                            the one that didn't need pam_selinux, and to drop
                            pam_selinux in openssh.

    star, procps, util-linux, shadow, vixie-cron:

            status: although patches are available from
                    http://www.nsa.gov/selinux/code/download5.cfm,
                    no bug-report or integration into debian/selinux have
                    been initiated for these packages.

            colin walters does have debian packages available
            (mirrored at http://selinux.lemuria.org/walters)

    login:

            status: what used to be a patch in login can be achieved
                    equally well with pam_selinux.so session.

            TODO: must write patch for kdm's /etc/pam.d/kdm to have
                    pam_selinux.so session required

    kdm:

            status: patch created to do context switch but due to the
                    design of kdm's backend the use of pam_selinux.so
                            session achieves the same goal, making patching kdm
                            unnecessary.
                            
            TODO: must write patch for kdm's /etc/pam.d/kdm to have
                    pam_selinux.so session required

    wdm:

            status: patch created but not yet accepted upstream. code
                    in wdm needs to be evaluated to see if pam_selinux.so
                            session will do the same job.

    gdm:

            status: patch accepted upstream to do session management.
                    it was essential in gdm that this be done because
                            the process doing authentication is separated from
                            the process doing the program running: pam_selinux.so
                            session would therefore be insufficient [without a
                            rewrite of gdm?]

    xdm:

            status: not known [to me].

    libselinux:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=251749

            status: still at priority "optional". 30may2004 message sent
                    to debian-devel requesting assistance in alerting
                            the "ftpmasters" to the issue. response: russell
                            should have received a notification because ftp.debian.org
                            automatically "overrides" the priority.

    postfix:

            http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=253732

            status: the postfix policy requires that you disable chrooting
                    in order for postfix to work. 253732 is a wish-list
                            requesting an extra dpkg config question advising people
                            to select "no i do not want to chroot" if they are
                            installing on an se/linux system.

    -- 
    -- 
    Information I post is with honesty, integrity, and the expectation that
    you will take full responsibility if acting on the information contained,
    and that, should you find it to be flawed or even mildly useful, you
    will act with both honesty and integrity in return - and tell me.
    --
    <a href="http://lkcl.net">      lkcl.net      </a> <br />
    <a href="mailto:lkcl@lkcl.net"> lkcl@lkcl.net </a> <br />
    -- 
    To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org 
    with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
    

  • Next message: Chris Metzler: "Re: Gnome 2.6 and MAC OSX-like panel"

    Relevant Pages

    • MS binary integrity baseline
      ... Is there anything that performs binary integrity checks for Windows OS such ... as rpm does for Redhat or apt does for Debian? ... I want something that will check Windows binaries against a trusted source - ...
      (Focus-Microsoft)
    • samba/pam/winbind username case
      ... I've got my debian terminal server authenticating from ... directories on the debian machine with pam mk_homedir. ... JSMITH and jsmith, which is fine, except it sometimes ... case accounts and probably eventually 600 duplicate ...
      (Debian-User)
    • [Full-disclosure] [SECURITY] [DSA 2326-1] pam security update
      ... Kees Cook of the ChromeOS security team discovered a buffer overflow ... PAM stack, which allowed the execution of arbitrary code. ... The oldstable distribution is not affected. ... Further information about Debian Security Advisories, ...
      (Full-Disclosure)
    • [SECURITY] [DSA 2326-1] pam security update
      ... Kees Cook of the ChromeOS security team discovered a buffer overflow ... PAM stack, which allowed the execution of arbitrary code. ... The oldstable distribution is not affected. ... Further information about Debian Security Advisories, ...
      (Bugtraq)
    • Re: Systemeditor aendern?
      ... also meist im Kontext von 'login' oder irgendwelchen ... Authentifizierungsgeschichten, sonst nicht. ... Selbst cron verwendet hier unter Debian per default pam_env.so. ... Leider verwendet ssh beim Remote-login mit Schlüssel kein PAM, ...
      (de.comp.os.unix.apps.misc)